Factors predicting pain and effect of oral analgesia in topical photodynamic therapy

Abstract

To The Editor, Topical photodynamic therapy (PDT) is highly efficacious for the treatment of actinic keratosis (AK), Bowen’s disease (BD) and superficial basal cell carcinoma (sBCC), and efficacious in thin nodular BCCs (nBCC) (1). It is generally well tolerated, with pain or discomfort being a known side effect (1). Pain is typically described as a burning or stinging sensation of varying intensity. Prior studies looking at factors which influence pain during PDT have shown conflicting results (2–5). The aim of our study was to identify any patient or treatment factors associated with higher levels of pain during PDT and to assess the effect of pretreatment analgesia. Treatment is standardised in our centre and documented in a standard treatment proforma. Patients receive one treatment per course for AK and two treatments per course, one week apart, for BD and BCC. Methyl aminolevulinate (MAL–Metvix Photocure ASA, Oslo, Norway) is used as a photosensitising agent. Wood’s lamp fluorescence (WLF) is used to assess absorption of the photosensitiser (excellent/good/poor). The Aktilite (Photocure ASA) (LED red light, 630 nm, 37J/cm) or Waldmann 1200L (Waldmann Medizintechnik, Villingen-Schwenningen, Germany) (filtered halogen lamp, non-coherent red light, 570–730 nm, 50J/cm) light-source is used. Patients complete a visual analogue pain scale (VAPS) to grade their pain during and after treatment. The highest pain score experienced during treatment is documented (range 0–100). Patients who are unable to complete a VAPS, for example, due to eyesight difficulty, are asked to grade their pain on a similar scale. Patients are offered oral analgesia 30 min pretreatment, and type of analgesia taken is documented. A fan and cooling water spray are provided during treatment. Following ethics approval, retrospective chart review was completed. Data were compiled in Microsoft Excel and analysed using SPSS (IBM SPSS Statistics for Macintosh, Version 22.0. Armonk, NY: IBM Corp.). All AK, BD and BCC treated between 2009 and mid-2015 were included in analysis. Complete data sets were available on all patients. Pearson’s chi-square, independent samples t-test and one-way ANOVA were used to obtain Pvalues, unless otherwise stated. Odds ratio (OR) and 95% confidence intervals (CI) were calculated using binary logistic regression. For the purpose of binary logistic regression, a dichotomous pain response was defined by dividing the VAPS into two groups at the median 40 (1–40 and 41–100). In total, 200 cases of AK, BD and BCC, in 109 patients were analysed. The average age was 74.6 years (Standard Deviation [SD] 9.55 range 33–96). The femaleto male-percentage split was 61 : 39. Sixty nine per cent of lesions were located on the limbs, 17% on the trunk and 14% on the head/neck. Fifty four per cent of lesions were BD, 24% BCC and 22% AK.